On Wed, Nov 19, 2014 at 10:48 PM, Greg Wooledge <wool...@eeg.ccf.org> wrote: > > No, I don't.
I completely respect that. > > If I'm writing in C, and I want to know whether one of my variables is > "empty", I have to know what type of variable it is. The code to check > whether a string (char x[10]) is "empty" is very different from the code > to check whether an array of integers (int x[10]) is "empty". Hell, > the meaning of "empty" would have to be defined by the application. > Would it mean every element is 0? Or something else? And what if x is > a float, or a double? Or a pointer? Or a struct? exactly. why is that? you touched the most important thing: types. thats why in script languages, where you do not define variables with types as such, you have the meaning of "empty", a="" makes a empty, declare -A b makes b empty. > > The same thing happens in bash. It's simply not the kind of language > where you have one syntax that works for multiple situations. well, I could argue:) but I wont. I see your point. cheers, pg
